Anirudha Majumdar
@Majumdar_Ani
Followers
3,993
Following
530
Media
122
Statuses
316
Assistant Professor in Robotics @Princeton . Visiting researcher @GoogleDeepMind in Princeton.
Princeton, NJ
Joined March 2020
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
RIIZE
• 119988 Tweets
Noa Argamani
• 112372 Tweets
Hobi
• 98216 Tweets
Nuseirat
• 92601 Tweets
THANK YOU BTS
• 92179 Tweets
مخيم النصيرات
• 87382 Tweets
#yks2024
• 62734 Tweets
Shlomi Ziv
• 60913 Tweets
Andrey Kozlov
• 60373 Tweets
Almog Meir Jan
• 55117 Tweets
SULWHASOO MOMENT GULF
• 46714 Tweets
サイバー攻撃
• 45070 Tweets
#世にも奇妙な物語
• 37151 Tweets
#BlankTheSeriesSS2EP4
• 33905 Tweets
#ネーションズリーグ
• 33509 Tweets
ULTIMATE LOVE FOR BLANK
• 32557 Tweets
バレーボール
• 31130 Tweets
マッサマン
• 30138 Tweets
#tyt2024
• 28879 Tweets
烈火の炎
• 26626 Tweets
SHOW STOPPER ELVISH
• 25976 Tweets
キスマイ
• 24993 Tweets
スロベニア
• 19143 Tweets
PITANDA NO E DE CASA
• 15455 Tweets
ジェシー
• 14133 Tweets
メインキャラクター
• 12547 Tweets
東キヨドーム
• 11481 Tweets
تحرير الاسري
• 10921 Tweets
男子バレー
• 10118 Tweets
Pinned Tweet
I received tenure at
#Princeton
! It’s been an incredible journey, and I am immensely grateful to my group, collaborators, mentors, and family. Looking forward to continuing our work on making 🤖 more trustworthy!
I will be on sabbatical 2024-25, spending time at
@GoogleDeepMind
.
48
3
449
📢 All 24 lecture videos for our Intro to
#Robotics
course
@EPrinceton
are now online!
All other course materials (lecture notes, slides, assignments, jupyter notebooks, final project code):
Hope someone finds these useful! Links👇
17
239
919
I am continuing to make all materials for our "Introduction to Robotics" course
@EPrinceton
openly available. New this Fall: video lectures!
All other materials (notes, slides, code, etc.) are available here:
9
165
813
With seemingly endless progress in AI, I decided over the past few months to take a deep dive into the state of robotics 🤖, emerging trends, and research challenges. What do actual field-deployed robotic systems look like now? Will LLMs solve robotics?
🧵
11
66
299
How can lightweight drones without depth cameras navigate using monocular images? Check out our paper at ISER 2023!
MonoNav: MAV Navigation via Monocular Depth Estimation and Reconstruction
arXiv:
website:
Work led by
@Nate___Simon
2
51
270
Up next in our Intro to
#Robotics
course at Princeton: vision & learning!
First lecture video in this module (intro to vision) is up:
Videos for estimation (Kalman+particle filtering), localization, and mapping are also online:
4
54
232
Thrilled that our paper on Robots That Ask For Help received the Best Student Paper Award at
#CoRL2023
! Congratulations to students
@allenzren
,
@anushridixit111
, Sasha Bodrova, & our wonderful collaborators at
@GoogleDeepMind
!
13
6
171
Interested in learning more about how
@BostonDynamics
uses model predictive control to achieve agile locomotion? Check out Scott Kuindersma's recent talk at our
@EPrinceton
Robotics Seminar:
Taskable Agility: Making Useful Dynamic Behavior Easier to Create
2
31
167
We are half-way through our Intro to Robotics course at Princeton, and the first 12/24 lecture videos are available (feedback control, motion planning, and Bayesian filtering):
Up next: localization, mapping, vision, and learning.
0
40
162
Our collaboration with
@GoogleDeepMind
on language-instructed robots that know when they don't know (KnowNo) was highlighted by
@EPrinceton
in this news article:
Check it out to learn more!
4
15
122
Our group will be presenting 3 papers at
#CoRL2023
next week! Stop by to learn about our latest work, and to meet the fantastic students who led this work.
1. (Oral) Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners
2
13
111
My first (and last?) paper where I am both first and last author will be published at
#ICML2023
:
Fundamental Tradeoffs in Learning with Prior Information
How much prior knowledge does one require in order to learn quickly on a given task?
🧵
1
10
108
We've had some fantastic speakers at our Princeton Robotics Seminar, and have started making the videos available here!
First up, Russ Tedrake on Diffusion Policies:
See here for the full list of speakers:
2
17
107
I am delighted to be one of the recipients of the ONR Young Investigator Program award this year!
The award will support our group's work on robot learning with formal guarantees on out-of-distribution detection and generalization.
5
0
106
I am delighted to have received the Sloan Fellowship in Computer Science! I'm very grateful to my research group, mentors, collaborators, and colleagues for making this possible. Great to see six recipients from
@Princeton
!
#SloanFellow
We are delighted to announce the winners of this year’s Sloan Research Fellowship! These outstanding researchers are shining examples of innovation and impact—and we are thrilled to support them. Meet the winners here: 🎉
#SloanFellow
#STEM
#ScienceTwitter
13
61
411
10
3
102
Thrilled that my PhD student Allen Z. Ren
@allenzren
was awarded the Jacobus Fellowship -- Princeton's highest honor for PhD students! He is one of 4 students to receive the honor across the university. Congratulations
@allenzren
!!
4
4
103
Just wrapped up the final projects for our Intro to Robotics course at
@EPrinceton
: vision-based
#drone
navigation in unknown obstacle fields (with the Crazyflie drones from
@Bitcraze_se
). Starter code is here:
Hardware details:
2
8
93
How can we rigorously quantify the uncertainty of pretrained perception models to provide end-to-end statistical safety assurances for robot navigation?Perceive With Confidence: Statistical Safety Assurances for Navigation with Learning-Based Perception
🧵
3
13
83
Interested in rigorous uncertainty quantification for neural networks? Check out our
#NeurIPS23
paper on combining conformal prediction and PAC-Bayes theory to obtain statistical guarantees on coverage and efficiency!
Paper:
Code:
3
13
82
Wrapped up our final project for Intro to Robotics at
@EPrinceton
: vision-based navigation and target-finding. I have made all course materials (slides, notes, assignments) available online:
The course covers control, planning, SLAM, and vision+learning.
4
14
81
So much to celebrate with our group! Vince Pacelli defended his PhD, Anushri Dixit (postdoc) accepted a faculty position at UCLA, and 5 undergrads are off to PhD programs (Michelle Ho:Stanford; Abhinav Agarwal:MIT; Neil Hazra:Columbia; Naomi Oke:CMU; Kristen Ahner: CU Boulder)!
1
2
77
Getting ready to teach Intro to Robotics again at
@EPrinceton
! I will continue to update course materials (slides, notes, assignments, code), which are all freely available here:
Video lectures from last year are on YouTube!
0
7
76
Join us on Monday at
#CoRL2023
for the workshop on Out-of-Distribution Generalization in Robotics!
We have a fantastic line-up of speakers and panelists:
@AnguelovDrago
(Waymo),
@chelseabfinn
(Stanford),
@yayitsamyzhang
(UT Austin),
@judyfhoffman
(GA Tech)
1
13
74
I am happy to have joined the new editorial board of the International Journal of Robotics Research (
#IJRR
) as an Associate Editor. I'm looking forward to helping the journal increase its impact, review quality, and timeliness in this new phase!
4
2
64
Very excited to celebrate the first PhD student to defend from my group at Princeton! Alec Farid: "Provably safe learning-based robot control via anomaly detection and generalization theory".
Alec is now joining the ML Planning team at
@zoox
!
0
2
55
Excited to have 2 papers accepted to
#RSS2022
.
1. Fundamental Performance Limits for Sensor-Based Robot Control and Policy Learning
We develop information-theoretic techniques for establishing fundamental limits on performance imposed by a robot's sensor.
2
3
53
I had a wonderful visit to the Univ. of Washington
@uwcse
to meet old friends and make some new ones!
Check out the talk here about our work on language-instructed robots that know when they don't know:
Thanks very much
@abhishekunique7
for hosting!
0
5
51
Delighted to have received a Young Faculty Researcher award from the Toyota Research Institute. Very excited to be collaborating with
@ToyotaResearch
on our efforts on learning manipulation with guarantees on distributional robustness/generalization.
3
1
51
Paper accepted to
@corl_conf
!
Leveraging Language for Accelerated Learning of Tool Manipulation
We investigate how language descriptions of a tool (e.g., geometry, common uses) can help a robot learn how to use the tool more quickly. Collab w
@karthik_r_n
3
4
51
I am teaching Intro to Robotics
@EPrinceton
this Fall and am making the course materials (slides, notes, assignments) available online:
The course is targeted at undergraduates (juniors/seniors) and has a graduate track too.
1
4
46
New preprint:
Comparing the Complexity of Robotic Tasks
We propose a definition of reduction between robotic tasks in order to capture when one task is harder than another, and also define a measure that quantifies the relative complexity between tasks.
1
5
46
How can we safely bridge the sim2real gap in vision-based navigation tasks?
Check out our paper, Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees, now accepted to the Artificial Intelligence Journal!
Website:
🧵
1
10
45
Getting ready to teach Intro to Robotics
@EPrinceton
. The course provides an intro to the fundamental theoretical and algorithmic principles behind robotic systems. The course also allows students to get hands-on experience using the
@Bitcraze_se
Crazyflie
#drones
.
1
8
43
Check out our new paper surveying applications, challenges, and the future of Foundation Models in robotics!
Foundation Models in Robotics: Applications, Challenges, and the Future
paper page:
We survey applications of pretrained foundation models in robotics. Traditional deep learning models in robotics are trained on small datasets tailored for specific
6
156
605
1
6
43
We have a newly minted PhD student from our group at
@Princeton
! Congratulations to Dr. Meghan Booker, and good luck on your journey as a researcher at
@JHUAPL
! 🎓🎓🎓
2
1
45
We are continuing to make the videos from our
@Princeton
Robotics Seminar available online!
Check out Henny Admoni's (
@hadmoni
) talk here:
Robots that Learn From and Collaborate with People
0
9
42
We hosted Changliu Liu from
@CarnegieMellon
for our
@Princeton
Robotics Seminar! Check out her talk on safe control of robots here:
Ensuring Robot Safety Through Safety Index Synthesis
0
6
41
Interested to learn more about our recent work on LLM planning using conformal prediction to ask for human help and provide statistical guarantees?
I will be presenting at the
#RSS2023
workshop on Safe Autonomy on Monday.
0
5
40
Our work on leveraging language for accelerated learning of robotic tool manipulation (led by PhD student
@allenzren
in collab. with
@karthik_r_n
) was featured in a new Scientific American
@sciam
article alongside a number of exciting advancements!
1
6
39
We just released code for MonoNav: MAV Navigation via Monocular Depth Estimation and Reconstruction!
Check it out here:
We just released code for MonoNav: MAV Navigation via Monocular Depth Estimation and Reconstruction! (w/
@Majumdar_Ani
at IRoMLab)
MonoNav enables real-time 3D reconstruction from a stream of noisy images and poses.
Code:
1
0
5
0
5
39
Excited to be heading to
#ICML
soon! I'll be presenting "Fundamental Tradeoffs in Learning with Prior Information" on Wed:
On Fri, I will present our group's work on generalization theory for robot learning at the PAC-Bayes workshop
1
4
38
I recently gave a talk at our own Mechanical & Aerospace Engineering department at
@EPrinceton
! Check it out for an overview of our work over the past few years on "Robots that know when they don't know":
0
6
38
Is your robot fundamentally limited in its safety or performance due to its sensor? How can we compute such fundamental limits 🛑?
Check out our paper, accepted to
#IJRR
!
1
6
36
For our course MAE 345/549: Intro to Robotics at
@EPrinceton
, we are shipping robot kits to each student with the
#JetBot
from
@NVIDIAAI
and
@waveshare00
. The course covers control, motion planning, SLAM, vision, and learning.
1
1
36
Getting ready for our first hardware assignment for MAE 345/549 (Intro to Robotics) at
@EPrinceton
using the
@nvidia
#JetBot
from
@waveshare00
. The robot runs
#cuda
and is capable of running neural networks on the camera feed. Excited to see the students use it!
0
0
36
New preprint in collaboration with Jaime Fisac's group:
Sim-to-Lab-to-Real: Safe Reinforcement Learning with Shielding and Generalization Guarantees
Work led by Kai-Chieh Hsu and Allen Ren.
2
1
34
LLMs are amazing at generating long-horizon plans. But they can also hallucinate outputs, which can be unsafe for robots! How can we build LLM planners that 𝑘𝑛𝑜𝑤 𝑤ℎ𝑒𝑛 𝑡ℎ𝑒𝑦 𝑑𝑜𝑛'𝑡 𝑘𝑛𝑜𝑤🤷 and ask for help to provide statistical guarantees?
LLMs can generate plans and write robot code 📝 but they can also make mistakes. How do we get LLMs to 𝘬𝘯𝘰𝘸 𝘸𝘩𝘦𝘯 𝘵𝘩𝘦𝘺 𝘥𝘰𝘯'𝘵 𝘬𝘯𝘰𝘸 🤷 and ask for help?
Read more on how we can do this (with statistical guarantees) for LLMs on robots 👇
4
47
236
0
8
34
Very excited to celebrate the successful thesis defense of Meghan Booker! Her thesis is titled:
Task-Driven Perception and Control for Robust and Efficient Autonomy
Meghan is off to
@JHUAPL
after Princeton! Congratulations Meghan!!
0
0
32
Celebrating our time with
@anushridixit111
, who has spent the year as a postdoc in our lab. She is heading off to
@UCLA
as an Asst. Professor starting July! If you are interested in robot control or related topics, keep an eye out for the amazing work from her new group!
0
0
32
Our group is presenting 2 papers next week at
#RSS2022
(Wed, Spotlights 6)! I will also be presenting at the RSS workshop on Risk-Aware Decision Making on Mon. Join us to learn more about our group's recent work. Excited to attend another in-person conference!
0
0
29
We have 3 papers from our group accepted to the Learning for Dynamics & Control (
#L4DC
) conference.
1. (Oral presentation) "Generating Adversarial Disturbances for Controller Verification" (w/ Udaya Ghai, David Snyder,
@HazanPrinceton
)
1
2
28
I decided to probe
#ChatGPT
’s ability to perform basic physical reasoning. I don’t think I’ll be using it to generate notes for my Intro to Dynamics course just yet.
Some fun examples 🧵
2
2
28
Our work on language-instructed robots that know when they don't know (KnowNo) was highlighted by MIT Tech Review
@techreview
in this news article!
0
3
28
But it still seems that many of the challenges in “lower-level” parts of the robotics stack (robust perception, control, planning w/ dynamic constraints, human-robot interaction) continue to remain challenging, especially when one desires low levels of failure in novel scenarios.
1
4
28
Excited to be presenting at the
@Princeton
Workshop on Optimization, Learning, and Control (
#OLC24
)📢June 27-28, 2024🗓️!
📢📢📢
@Princeton
Workshop on Optimization, Learning, and Control (
#OLC24
)📢📢📢
Mark your calendars🗓️ June 27-28, 2024
We will have a poster session for junior researchers (submission deadline March 3)!
🔗
#optimization
#learning
#ml
#control
#orms
2
25
142
0
6
27
We have 3 papers from our group accepted to
@corl_conf
, including one oral presentation! Congratulations to students
@allenzren
(x2), David Snyder,
@anushridixit111
, Meghan Booker, Sasha Bodrova,
@Nate___Simon
, and our collaborators!
Links to papers 👇
2
1
26
New pre-print from our group: "Task-Driven Out-of-Distribution Detection with Statistical Guarantees for Robot Learning".
We utilize PAC-Bayes theory to learn policies with bounds on performance. Violation of the bound implies OOD with high confidence.
2
0
26
Paper accepted to the Conference on Robot Learning (
#CoRL2021
)!
"Task-Driven Out-of-Distribution Detection with Statistical Guarantees for Robot Learning"
ArXiv:
Work led by: Alec Farid and
@Veer_Sushant
1
0
26
Hey robot, where did I leave the books I was reading? Check out our work on embodied question answering via VLM-based semantic reasoning and conformal prediction! Collaboration with
@DorsaSadigh
and
@ToyotaResearch
.
👇🧵
How do we get robots to efficiently explore diverse scenes and answer realistic questions? e.g., is the dishwasher in the kitchen open❓
👇Explore until Confident — know where to explore (with VLMs) and when to stop exploring (with guarantees)
1
11
66
0
2
26
Paper from our group accepted to
#WAFR2022
:
Towards a Framework for Comparing the Complexity of Robotic Tasks
Work led by Michelle Ho (undergrad student in our group!) and Alec Farid (PhD student).
1
0
25
We are presenting two papers at
#CoRL2020
this Wed on making generalization and safety guarantees for learning-based planning/control for robots. Join
@Veer_Sushant
and
@allenzren
at Parallel Interactive sessions II and III on Wed to learn more!
0
2
25
The MAE department
@EPrinceton
has multiple faculty position openings at the assistant, associate, or full professor ranks, with Robotics as one of the focus areas:
Consider applying if you are looking at academic positions this cycle!
0
2
24
I gave a talk on our work "Robots that ask for help: conformal prediction for LLM planners" at the International Seminar on Distribution-Free Statistics organized by
@ml_angelopoulos
:
Check out the talk to learn more about this recent work!
1
5
24
Join our growing efforts in
#Robotics
at
@EPrinceton
! We have a new faculty position in robotics at junior/senior levels:
Link:
#AcademicTwitter
#AcademicChatter
0
2
24
Our group is presenting 3 papers next week at
#ICRA2022
in the following sessions:
Tue: TuB17, Room 123
Wed: WeA06, Room 115A
Wed: WeA19, Room 124
Join us to learn more about our recent work and to meet some fantastic students! Details below.
1
0
23
Very excited to finally have an in-person event with all the PhD students/postdocs (+postdog) in our group!
0
2
24
Presenting two collaborative projects with
@HazanPrinceton
's group at the
@MSFTResearch
Reinforcement Learning Day on Jan 14:
1. Generating Adversarial Disturbances for Controller Verification
2. A Regret Minimization Approach to Iterative Learning Control
2
2
23
We are looking for outstanding postdoc candidates with research interests that broadly overlap with our group (safety and performance guarantees for
#robots
; guarantees for
#learning
-based control; robust
#perception
and
#control
). More information here:
4
3
23
I had a great time presenting our group's work at the ETH Autonomy Talks series. The talk video is available here:
Safety and Generalization Guarantees for Learning-Based Control of Robots
0
2
22
Interested in out-of-distribution generalization in robotics? We are organizing a workshop at
#CoRL2023
, with an amazing lineup of speakers and panelists!
📢 Announcing the first
@corl_conf
workshop on Out-of-Distribution Generalization in Robotics: Towards Reliable Learning-based Autonomy!
#CoRL2023
🎯 How can we build reliable robotic autonomy for the real world?
📅 Short papers due 10/6/23
🌐
🧵(1/4)
1
15
33
0
6
21
For the final project for our course "MAE 345/549: Intro to Robotics" at
@EPrinceton
, students trained
#NeuralNetworks
to perform vision-based obstacle avoidance and navigation on the JetBot from
@NVIDIAAI
and
@waveshare00
. Wonderful to see the great work from the students!
1
1
21
Our work on making safety/generalization guarantees for
#learning
-based control of
#robots
is highlighted in a
@EPrinceton
news article. Work led by Alec Farid,
@allenzren
,
@Veer_Sushant
, and Anoop Sonar.
A small drone successfully takes flight through a room filled with randomly placed cylinders acting as stand-ins for trees, people or structures.
How? Machine learning.
Step inside
@Majumdar_Ani
's Intelligent Robot Motion Lab to learn more:
0
4
17
1
2
21
Very happy to be involved with this work on an open-source differentiable control library w.
@HazanPrinceton
's group.
Tech report:
Code:
Happy to announce our open-source differentiable control library w. many of our new algorithms for non-stochastic control!
Talk by my student Paula Gradu Friday 12:23 - 12:35 EST, poster: 16:30 - 17:30 EST
3
20
142
1
0
21
Announcing
#ICRA
2023 Workshop:
Bridging the Lab-to-Real Gap: Conversations with Academia, Industry, and Government (Jun 2, London).
We have a fantastic lineup of invited speakers!
Please consider submitting an abstract (Apr 14):
1
2
21
Our work on leveraging language models for accelerated learning of robotic tool manipulation (led by PhD student
@allenzren
in collab. with
@karthik_r_n
) was featured in an article from
@QuantaMagazine
alongside a number of exciting recent advancements:
0
3
20
@HazanPrinceton
gave a talk at our
@EPrinceton
Robotics Seminar on the theory of online learning and its applications to robotics! Check out the talk here:
We will continue to post our seminars online!
0
6
20
Our group is presenting 3 papers today at
#L4DC
. David Snyder is presenting work w/ Udaya Ghai and
@HazanPrinceton
in Oral Presentations 1B (Adversarial Learning). Meghan Booker, Anoop Sonar, and Vince Pacelli are presenting in Poster Session 1A.
Join us!
0
0
20
We have a really exciting program for our
#ICRA
workshop on "Bridging the Lab-to-Real Gap", with speakers from
@Waymo
(Matt O'Kelly),
@SkydioHQ
(Hayk Martiros),
@NVIDIAAI
/ Stanford (
@drmapavone
), U. Michigan (
@necozay
), UCSD (
@SylviaLHerbert
), and UK civil aviation (A. Berry).
1
4
20
We are looking for outstanding
#postdoc
candidates for our research group at Princeton:
Candidates broadly interested in safety and performance guarantees for robots are welcome to apply. More information here:
0
4
19
Excited to have 2 papers accepted to
#RSS2024
! Congrats to PhD students
@justinlidard
and
@allenzren
for leading the work!
Papers👇
1
3
19
Delighted to have our group's research featured in the A.I. Nation Podcast from
@whyy
and
@Princeton
(Ep. 2: AI in the driver's seat):
Thanks to
@malcolmburnley
,
@alexraestern
, and
@EdFelten
for highlighting work from
@Veer_Sushant
and others in our group.
1
2
18
How can we improve drone performance in gusty conditions?
Check out our paper at
#ICRA2023
!
FlowDrone: Wind Estimation and Gust Rejection on UAVs Using Fast-Response Hot-Wire Flow Sensors
🧵
2
2
16
Wonderful to see the different approaches teams in our Intro to Robotics course
@EPrinceton
took to their final project (navigation with
@Bitcraze_se
drones)! Teams used NNs for monocular depth estimation, reactive controllers, and more.
Lectures:
1
1
18
There is much more in the write-up:
I wrote this mostly for myself and it is not meant to be an academic survey. But, I am making it accessible in case others find it useful as a source of further sources!
1
2
18
Our paper on “Generalization Bounds for Meta-Learning via PAC-Bayes and Uniform Stability” was accepted to
#NeurIPS2021
:
We derive a framework for providing generalization bounds by exploiting the particular structure of meta-learning.
Led by Alec Farid.
3
0
18
For my course MAE 345/549: Intro to Robotics at Princeton, I have been using
@googlejamboard
in combination with
#zoom
. Jamboard allows students to refer back to previous pages while I'm writing (e.g., refer back to equations, definitions, etc. from earlier on in the lecture.)
0
0
17
New pre-print on recent collaborative work:
FlowDrone: Wind Estimation and Gust Rejection on UAVs Using Fast-Response Hot-Wire Flow Sensors
We integrate novel MEMS flow sensors onto a drone, and train a policy using RL for real-time control with wind.
1
1
16
Two new preprints from our group, both with Princeton undergrad first authors!
1. Stronger Generalization Guarantees for Robot Learning by Combining Generative Models and Real-World Data.
Work led by Abhinav Agarwal,
@Veer_Sushant
, and
@allenzren
.
2
1
16
New collaborative work with
@karthik_r_n
's group:
Leveraging Language for Accelerated Learning
of Tool Manipulation
We investigate how language descriptions of a tool (e.g., geometry, common uses) can help a robot learn how to use the tool more quickly.
1
3
16
New pre-print from our group: "Distributionally Robust Policy Learning via Adversarial Environment Generation" (DRAGEN).
ArXiv:
Video:
Work led by
@allenzren
.
2
1
15
@Abhismirai10
There are interesting challenges w.r.t. distribution shift that arise when using pre-trained perception modules as part of a feedback controller/planner (specifically, novel viewpoint angles, etc.); there is some great discussion in this talk by
@rsalakhu
0
3
14
@hausman_k
@ericjang11
I also really like this talk from
@jpineau1
, where she makes a similar point by estimating information content (bits) in sim vs. reality:
0
0
14
We have 3 papers accepted to
#ICRA2022
!
1. Stronger Generalization Guarantees for Robot Learning by Combining Generative Models and Real-World Data
Work led by Abhinav Agarwal (undergrad student in our group!),
@Veer_Sushant
, and
@allenzren
4
2
15
A video from our Intro to Robotics course (MAE345)
@EPrinceton
last semester. The students programmed Crazyflie
#drones
from
@Bitcraze_se
to dodge PVC pipes using vision-based control.
0
4
15
Looking forward to participating as a panelist at the IROS Workshop on Safe Real-World Robot Autonomy! Join us (virtually) on September 27th for the workshop.
Join us on September 27 at the IROS Workshop on Safe Real-World Robot Autonomy! Check out our program and add your questions for the panelists here:
Early registration closes September 5!
0
4
22
0
0
15
18. Optical Flow
19. Intro to Deep Learning
20. Stochastic Gradient Descent
21. Multi-layer Networks, Overfitting, Regularization
22. Convolutional networks
2
0
14
New pre-print from our group: "PAC-BUS: Meta-Learning Bounds via PAC-Bayes and Uniform Stability".
We derive generalization guarantees for gradient-based meta-learning using PAC-Bayes and algorithmic stability.
Work led by Alec Farid.
1
0
14